End of training

README.md

@@ -4,6 +4,7 @@ license: llama3.1
 base_model: meta-llama/Meta-Llama-3.1-8B
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: llama3-1_8b_math_50000_samples
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3-1_8b_math_50000_samples
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/math_50000_samples dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6146
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.6146200895309448,
+    "eval_runtime": 4.7986,
+    "eval_samples_per_second": 231.527,
+    "eval_steps_per_second": 1.042,
+    "total_flos": 211054692925440.0,
+    "train_loss": 0.6023140767263988,
+    "train_runtime": 1093.8559,
+    "train_samples_per_second": 57.844,
+    "train_steps_per_second": 0.115
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.6146200895309448,
+    "eval_runtime": 4.7986,
+    "eval_samples_per_second": 231.527,
+    "eval_steps_per_second": 1.042
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 211054692925440.0,
+    "train_loss": 0.6023140767263988,
+    "train_runtime": 1093.8559,
+    "train_samples_per_second": 57.844,
+    "train_steps_per_second": 0.115
+}

trainer_state.json

@@ -0,0 +1,150 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 126,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 7.438439570937922,
+      "learning_rate": 5e-06,
+      "loss": 0.7676,
+      "step": 10
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 1.6813673118400363,
+      "learning_rate": 5e-06,
+      "loss": 0.6743,
+      "step": 20
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.9975184383465511,
+      "learning_rate": 5e-06,
+      "loss": 0.6533,
+      "step": 30
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.7160368762855558,
+      "learning_rate": 5e-06,
+      "loss": 0.6367,
+      "step": 40
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.6265256404876709,
+      "eval_runtime": 4.6089,
+      "eval_samples_per_second": 241.055,
+      "eval_steps_per_second": 1.085,
+      "step": 42
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 0.9365883424052448,
+      "learning_rate": 5e-06,
+      "loss": 0.6056,
+      "step": 50
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.5269751435450583,
+      "learning_rate": 5e-06,
+      "loss": 0.5894,
+      "step": 60
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.7088023568651957,
+      "learning_rate": 5e-06,
+      "loss": 0.588,
+      "step": 70
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.7190623672748503,
+      "learning_rate": 5e-06,
+      "loss": 0.5825,
+      "step": 80
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6075287461280823,
+      "eval_runtime": 4.4883,
+      "eval_samples_per_second": 247.533,
+      "eval_steps_per_second": 1.114,
+      "step": 84
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 0.8144598049817143,
+      "learning_rate": 5e-06,
+      "loss": 0.5563,
+      "step": 90
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.7353847410203146,
+      "learning_rate": 5e-06,
+      "loss": 0.5354,
+      "step": 100
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.785281004963026,
+      "learning_rate": 5e-06,
+      "loss": 0.5417,
+      "step": 110
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.6504834481815525,
+      "learning_rate": 5e-06,
+      "loss": 0.5386,
+      "step": 120
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.6146200895309448,
+      "eval_runtime": 4.5411,
+      "eval_samples_per_second": 244.656,
+      "eval_steps_per_second": 1.101,
+      "step": 126
+    },
+    {
+      "epoch": 3.0,
+      "step": 126,
+      "total_flos": 211054692925440.0,
+      "train_loss": 0.6023140767263988,
+      "train_runtime": 1093.8559,
+      "train_samples_per_second": 57.844,
+      "train_steps_per_second": 0.115
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 126,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 211054692925440.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}